Click here to view the embedded video.

It is pretty mesmerizing to see the absolute chaos that comes from such a tiny little bullet.  I like the ballistics gel shots the best.

This has to be one of the first “CAD Demonstrations” in history, and it is pretty amazing what could be done with 1963 computer technology (actually the rig they are on was built in 1958, predating the microchip).  From the description, this video depicts:

…the software Ivan Sutherland developed in his 1963 thesis at MIT’s Lincoln Labs, “Sketchpad, A Man-Machine Graphical Communication System”, described as one of the most influential computer programs ever written. This work was seminal in Human-Computer Interaction, Graphics and Graphical User Interfaces (GUIs), Computer Aided Design (CAD), and constraint/object-oriented programming.

Here are both parts:

Click here to view the embedded video.

Click here to view the embedded video.

They cover all of the basics for 3D computer modeling.  They have sketching, snapping, constraining, modifying, importing, and view manipulation down just as well as any modern CAD software (minus some of what amounts to special effects).  In fact, the stylus interface used is only recently coming back into style.  And there is some discussion about other human-computer interfaces they were developing at the time that we still haven’t quite figured out, like speech and handwriting recognition.  I guess hard problems are hard.

It’s particularly interesting to me the blend of parametric and direct modeling, being able to apply rigid constraints on the one hand and deforming the model dynamically on the other.  It seems like MIT Lincoln Labs had CAD all figured out a half century ago, and the rest has all been largely cosmetic.  You can even find Ivan Sutherland’s entire thesis on Sketchpad online here.  In it he details everything from constructing the “light-pen” to the different constraint solving methods available.

I especially like the first demonstrator’s objection handling.  Like in the second video (5:45) when asked about drawing non-linear models, such as an automobile, and he simply replies, “we are well on our way with surfaces.”  Brilliant.  I think I’ve used that exact line a dozen times, 46 years later.

I think it’s especially important to note how much enjoyment everyone seems to get out of this little experiment.  The guys on stage are smiling, the audience laughs as they play along, and now hundreds of thousands of folks are watching online.  People like things they are familiar with, and it is striking when something they have never actually experienced before hits that familiarity chord anyways.  There is an intrinsic attraction to consonance, and maybe some muster behind the notion that music is a universal language.

I like this picture. I took this picture. Now I’m using it to see if my Flickr account is actually linked correctly to Jabberwocky.

I work as an Application Engineer for SpaceClaim, which means I spend a large portion of my waking hours using the software. Which, it turns out, allows me to create a model with far more detail that is probably FDA approved. I began work on a scale model of a dragonfly last August as a response to some of the new functionality we were introducing into SpaceClaim 2009. Here is what I ended up with:

Almost Finished Dragonfly

I learned a lot about SpaceClaim while building this little bug, so I figured I could share the experience and maybe show others what it can do. And possibly stroke my ego a bit while I’m at it.

The Idea

When I started, I was thinking more of a steampunk style insect. I had probably seen a picture on BoingBoing that I thought looked neato, and I remember searching Wikipedia for insect wings one afternoon. I found these:

Front View of Wings

Close-up of Wing Joint

After finding these images it didn’t take me long to drop the steampunk idea. I figured that the actual natural forms would be an interesting change of pace; I have already built a million boxy designs in SpaceClaim.

The Wing

From here I began with the wing. This was easily the most tedious part of the project. I traced somewhere between a lot and a ton of intricate little wing veins. I imported the image right into my design, and set the scale to be around 3 inches (it’s a BIG dragonfly). Using layout curves allowed me to switch back to 3D without forming surfaces. I organized the veins into four rough classes, by thickness, and gave each a layer (note the “layoutPlane” layer, used to hide the datum plane but not the curves). I could then set lineweights and colors to help visualize the wing structure early on.

Layers for Wing

I carefully traced every tiny vein in the wings using lines and arcs. It was around this time that Blake caught wind of what I was working on, and I think I nearly moved him to tears. While he may have been a little overdramatic, he did offer some good advice: too many picks in too short a time can lead to stress fractures (ask him about it). I took caution and paced myself, using the same process to create both the top and bottom wings, which would be mirrored in the final model. Here’s the general progression of the top wing:

At this point I had in the vicinity of 1000 curves drawn for each wing. It was unfeasible to create a cylinder or sweep for each curve by hand, but luckily Blake had just written an Add-in that did just that. His “Skeletize” tool allowed me to take the thousands of edges and create a solid skeleton from them (this functionality actually made its way into SpaceClaim proper, under the Cylinder tool). I broke the layout curves into smaller groups and projected them back into sketch curves, then went to 3D to create surfaces. SpaceClaim will only convert 200 sketch curves to surfaces in one go, but I wanted to have smaller surfaces to help with rendering later on anyway. After combining all the new solids I had two wings, each with one solid representing all of the veins and about eight surfaces forming the wing membrane.

Moving On – The Tail

With the easy but slow part done, it was time to dig into to the body, and I chose to start with the tail. Again, I went back to Wikipedia for a source image:

Side View of Tail

I blocked out the rough components very quickly using cylinders and spheres, just to get a sort of base to work with. I did some sketching where the primitives weren’t giving me enough, like the legs. This allowed me to structure my model, creating the sub-assemblies and layers I needed.

Moving into more detail, I began assembling the segments of the tail. I don’t really know the true physiology of dragonfly joints, so I took some liberties and simplified. I made simple slot-and-tabs with Center associations for each of the six segments.

Tail Joint Cross Section

I repeated this down the tail to the tip, which has some form of a pincer at the end. This I formed from some simple blends, then I applied similar assembly conditions as the existing segments.

To create some simple “veins” on the body, I chose to imprint the faces with intersecting datum planes, and then use the “Skeletize” tool again to create raised regions I could use the new Color Face commands on.

Slicing Up the Tail

Legs and Head and Body, Oh My!

I’ll come back to explain the process on creating the legs, head and body of the dragonfly, as well as assembling the whole thing later.

• Fixed the Archives to actually display past posts.
• Fixed the pagination links to actually fit the theme, and refined some of the locations they show up.
• Worked on showing multiple posts where they fit, and not where they don’t. I think I need to get Dusty to use the "<!--more -->" tag.
• Addressed some other minor styling issues.

The next thing to work out is the comment styling.  It looks a bit forgotten right now, because it sort of has been.

The class was broken down to teams of 3 or 4.  My group designed and constructed the gripper assembly, which had to hold the razor secure as it was dropped and then release it without perturbing it at the bottom.  I also worked on the fabrication of the chassis and electrical system.

Here’s the video showing the operation in action:

The first part shows the razor being positioned to a precise orientation.  It is then secured in the grip and lifted to the top of the drop track.  We had access to a high speed camera to get the footage at the end showing the razor dropping the last few inches without rotating or changing orientation.

It was one of the best courses I took at WPI.  We built something to solve an actual problem, from brainstorm to functioning device, in seven weeks.  Not to mention the task of managing that group to stay on task.  Hats off to Prof. Norton for the opportunity and mentoring.  Now if I can only remember some of that stuff over the next 7 weeks as I get ready for the FEs.

I’m not 100% convinced I like the blockquotes I came up with, but they are an improvement and functional.

Adding the social network links was simple EXCEPT for Facebook giving me a ridiculous run around.  It is pretty easy to come up with the URL linking to your public search profile; I found it on the Badges page.  What is more difficult to figure out is that my search listing was not activated, so I kept getting 404s on the link.  Once I finally found the damn checkbox buried in the privacy controls the whole thing worked fine.

I decided to use FeedBurner despite some recent hiccups with the service.  I’ve never had an issue with it, and the plug-ins are out there to use easily.  You can follow Jabberwocky Design’s feed here; I’m lumping everything together for the time being, but I might spllit into at least two feeds in the near future.  In addition I’m looking at Twitme to syndicate posts directly to Twitter.  This post will test that as well.

So with that, I think I can call this Vorpal 0.5.  I still need to fix the IE issues, but those should be quick to do, and IE is terrible anyways.  I’ve still got some big things to fix here, but I think things are on the mark as far as syndication and social-ness are concerned.